Showing results for 
Search instead for 
Did you mean: 

LIS2DE12 and a buzzer on board


Hi, folks,

I'm using LIS2DE12 on the PCB. The purpose: the device should inform the firmware, if the PCB has moved - intruder detection. I successfully programmed the device for INT1 on 6D movement. It works fine in general.

On the PCB I have also a small (10mm) piezo buzzer. As soon as the buzzer sounds, LIS2DE12 detects a movement. The distance between LIS2DE12 and the buzzer is about 3cm (1.2"). I watched the data from LIS2DE12 during the buzzer run and I saw really a wild dance on X/Y/Z axies.

Well, I do understand, that the buzzer causes the vibrations, but did not expect to see values +/- 0.4g and more!

The accelerometer has only a high pass filter (to exclude the DC components) but no low pass filter to cut off anything above 1kHz (actually, the buzzer runs at 4kHz - at its own resonance).

Do you have any ideas to avoid false positive detection while the buzzer is running?

P.S. reducing the sampling frequency (now at 400Hz) as a "low pass filter" does not help and makes the things even worse...

ST Employee

Hi @ColdWeather​ ,

It looks difficult to disentangle the contribution to vibration due to the buzzer with the LIS2DE12.

What I might suggest is to characterize at time 0 the noise related to the buzzer working on the board, and then "subtract" it (in the form of zero g level contribution or with an FFT postprocessing to the acquired signal.